signal handlers in a fork()ed child process when Python is compiled with
thread support. The bug was reported by Scott <scott@chronis.icgroup.com>.
What happens is that after a fork(), the variables used by the signal
module to determine whether this is the main thread or not are bogus,
and it decides that no thread is the main thread, so no signals will
be delivered.
The solution is the addition of PyOS_AfterFork(), which fixes the signal
module's variables. A dummy version of the function is present in the
intrcheck.c source file which is linked when the signal module is not
used.
in a few places -- so it can be set to .exe on GNUWIN32 platforms and
do the right thing. Whatever. (This was already done in
Modules/Makefile* but wasn't carried over here.)
All geometry manager methods that apply to a master widget instead of
to a slave widget have been moved to the Misc class, which is
inherited by all of Tk(), Toplevel() and Widget(). They have been
renamed to have their geometry manager name as a prefix,
e.g. pack_propagate(); the short names can still be used where
ambiguities are resolved so that pack has priority over place has
priority over grid (since this was the old rule).
Also, the method definitions in the Pack, Place and Grid classes now
all have their respective geometry manager name as a prefix
(e.g. pack_configure); the shorter names are aliases defined through
assignment.
A similar renaming has been done for all config() methods found
elsewhere; these have been renamed to configure() with config being
the alias (instead of the other way around). (This may not make much
of a difference but the official Tk command name is now 'configure'
and it may help in debugging tracebacks.)
Finally, a new base class BaseWidget has been introduced, which
implements the methods common between Widget and Toplevel (the
difference between those two classes is that Toplevel has a different
__init__() but also that Toplevel doesn't inherit from Pack, Place or
Grid.
sys.stdin.readline(), you get a fatal error (no current thread). This
is because there was a call to PyErr_CheckSignals() while there was no
current thread. I wonder how many more of these we find... I bnetter
go hunting for PyErr_CheckSignals() now...
immediately following colons. Sjoerd noticed this one too. Here's a
nonsense.py file that flexes all the font-lock keyword combinations.
class A:
class B(A):
pass
def __init__(self):
if i == 2 and j == 3 or k == 4:
import stuff
from otherstuff import cool
for i in range(cool.count):
if i == j:
break
elif j == 1:
continue
print i
else:
return not i
elif q is not i:
return lambda x: x + 1
else:
try:
try:
raise stuff.error
except stuff.error, v:
print v
except:
global q
finally:
while q > 0:
q = q - 1
assert q == 0
def make():
a = A()
exec "nonsense"
del a
on NTEmacs 19.34.6.
(py-serial-number): New variable.
(py-execute-region): If make-temp-name is broken, simply append a
serial number to the string "python-" to get a temporary file name.
It's possible concurrent NTEmacs can step on each others toes, but it
makes no sense to further coddle a busted NTEmacs.
(py-mode-map): Moved py-mark-def-or-class from M-C-h to C-c C-m since
the old binding conflicts with the standard global backward-kill-word
binding, and this new binding is more conformant with other language
modes. Moved py-mark-block to C-c C-k.
Setting interp->builtins to the __builtin__ module instead of to its
dictionary had the unfortunate side effect of always running in
restricted execution mode :-(
I will check in a different way of setting __main__.__builtins__ to
the __builtin__ module later.
Also, there was a typo -- a comment was unfinished, and as a result
some finalizations were not being executed.
In Bart Simpson style,
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
I Will Not Check In Untested Changes.
(py-electric-backspace, py-electric-delete): Support the XEmacs 20 Way
for backspace and delete mappings. In XEmacs 19, Emacs 19, and Emacs
20, both backspace and delete keysyms are bound to
py-electric-backspace. In XEmacs 20, backspace and delete keysyms are
bound separately, allowing the user to specify forward or backward
deletion of the delete keysym through the variable
delete-key-deletes-forward. All this is the Right Way To Do It and
this implementation was largely ripped from CC Mode.
(py-execute-file): Better interaction with comint. Set
comint-scroll-to-bottom-on-output to t. Wrapper buffer change in
unwind-protect in case process filter fails.
(py-shell): Start Python with -i flag to fix tty problem on Windows;
presumably -- not yet tested.
(py-clear-queue): New function to clear the pending exec file queue.
Not currently keybound.
(py-execute-region, py-execute-buffer): Added optional async flag (use
via C-u prefix) to execute the region in a new asynchrous buffer, even
if the Python shell is running.
(py-append-to-process-buffer): Removed as obsolete. Comint provides
this functionality.
Removed fbound test defun of match-string. All modern X/Emacsen have
this function.
will compare equal even if the master file uses only \n to terminate
lines (this is by far the most common situation). Also, check for the
case where the master file is missing, and print the time difference
in seconds when the slave file appears newer than the master (for
debugging).
methods. Using None causes problems if the destructor is called after
the __builtin__ module has already been destroyed (unfortunately, this
can happen!). I can't just delete the object because it is actually
tested for (if self._sock: ...). Setting it to 0 is a bit weird but
works.
to inside floatsleep(). This is necessary because floatsleep() does
the error handling and it must have grabbed the interpreter lock and
thread state before it can do so.
- The interp->builtins variable (and hence, __main__.__builtins__) is
once again initialized to the built-in *module* instead of its
dictionary.
- The finalization order is once again changed. Signals are finalized
relatively early, because (1) it DECREF's the signal handlers, and if
a signal handler happens to be a bound method, deleting it could cause
problems when there's no current thread around, and (2) we don't want
to risk executing signal handlers during finalization.
his *own* RCS file for python-mode.el, and I've agreed that it would
be better if his version was in the Python source tree. However I
don't want to totally get rid of the old RCS file (which has
interesting info such as which version was in which Python release).
So I've moved the old one to python-mode-old.el behind the scenes,
and this checkin message indicates that I'm now deleting it.
If you do an update, you will actually get Barry's *new* version!
Introductory comment updates.
(python-font-lock-keywords): Added "assert"
(py-block-closing-keywords-re): New variable.
(py-no-outdent-re): Rewrite to use py-block-closing-keywords-re.
(py-shell): py-process-filter should no longer be necessary. Comint
should do all the work. Note that more fixes to the py-shell process
mechanism need to be done.
(py-execute-region): Check for empty region. Some questionable
changes to set-buffer after shell-command-on-region. Again, this all
needs to be closely examined for X/Emacs 19/20 compatibility.
(py-goto-beyond-final-line): py-parse-partial-sexp-works-p should no
longer be necessary.
(py-statement-closes-block-p): Use py-block-closing-keywords-re.
__init__.py (or __init__.pyc/.pyo, whichever applies) is considered a
package. All other subdirectories are left alone. Should make Konrad
Hinsen happy!
in libmath.a so they are available to mathmodule.so (in case it is
shared). While this still gets triggered on Solaris 2.x, this appears
to be harmless there.
Guido van Rossum
Barry Warsaw